phenotype (Table 2). all Pin1 mutants, includ- ing Y23F, that did not bind phosphoproteins failed to support cell growth (Fig. 4). These results indicate that pSer- or pThr-binding ac- tivity of the Pinl 14% ' domain is required for the protein function in vivo. We have demonstrated a fi~nction of WW domains as pSer- or pTh-binding modules and an essential role for U5V domains in mediating protein-protein interactions. Serine or Thr phos- phoiylation, often on PEST sequences (rich in Pro, Glu, Ser, and Thr). conQols the timing of ubiquitination of various proteins, and ubiq- uitin-protein ligases are responsible for sub- strate recognition (4). Our results indicate that WU' doinains of ubiquitin ligases may bind pSer- or pTh-containing sequences, thus tar- geting their catalytic donlains to phosphoryl- ated substrates to initiate protein degradation. Proline residues can put luilks into polypeptide chains because they undergo cis or trans isome~ization catalyzed by PPIases (24). Phos- phorylation reduces the isomerizatioil rate of the pSer- or pThr-Pro bond, and Pinl is a uniq~~e enzyme that isomerizes the pSer- or pThs-Pro bond and regulates activity bf phos- phoproteins (12, 13). The Pinl catalytic domain neither interacts with protein substrates in vitro, nor perfo~lns the essential fimctioil of the pro- tein in vivo. The Mr domain binds pSer- or pThr-Pr~ontaining peptides and mediates Pin1 interactions with most substrates. One of its biological functions may be to facilitate the processive isome~ization of Pinl substrates. SH2 domains have similar functions. Some SH2 domains in Tyr lunases preferentially bind pTyr residues phosphorylated by the catalytic domain, thereby increasing the local kiilase concentration so that substrates are processively phosphorylated on nlultiple sites (25). Because Pinl substrates are regulated by phosphoryl- ation of multiple Ser or Thr residues clustered at the regulatory domains (12-14, 26). Isomer- ~zation of these sites may provide a means to generate coordinate "all-or-none" actikity of heavily phospho~ylated phosphoproteiils References and Notes 1. T. Pawson, Nature 373, 573 (1995); C. B. Cohen, R. Ren, D. Baltimore, Cell 80, 237 (1995). 2. T. Pawson and J. D. Scott, Science 278, 2075 (1997). 3. B. j. Mayer and D. Baltimore, Trends Cell Biol. 3, 8 (1993); T. Pawson and J. Schlessinger, Curr. Biol. 3, 434 (1993); J. Schlessinger, Curr. Opin. Genet. Dev. 4, 25 (1994). 4. M. Rechsteiner and 8. W. Rogers, Trends Biol. Sci. 21, 267 (1996); B. E. Clurman et al., Cenes Dev. 10, 1979 (1996); R. W. King et al., Science 274, 1652 (1996); I. M. Verma and J. Stevenson, Proc. Natl. Acad. Sci. U.S.A. 94, 11758 (1997). 5. A. J. Muslin et a/., Cell 84, 889 (1996); M. B. Yaffe et al., ibid. 91, 961 (1997). 6. M. Scheffner, U. Nuber, j. M. Huibregtse, Nature 373, 81 (1995); j. M. Huibregtse et al., Proc. Natl. Acad. Sci. U.S.A. 92, 2563 (1995); M. Hochstrasser, Curr. Opin. Cell Biol. 7, 215 (1995). 7. D. Rotin, Curr. Top. Microbiol. Immunol. 228, 115 (1998); M. Sudol, Prog. Biophys. Mol. Biol. 65, 113 (1996). 8. K. P. Lu, 8. D. Hanes, T. Hunter, Nature 380, 544 (1996). 9. M. J. Macias et al., ibid. 382, 646 (1996); R. Ranga- nathan et al., Cell 89, 875 (1997). 10. H. I. Chen and M. Sudol, Proc. Natl. Acad. Sci. U.S.A. 92, 7819 (1995); 0 . Staub et a/., EMBO j. 15, 2371 (1996): M. T. Bedford, D. C. Chan, P. Leder, ibid. 16, 2376 (1997). 11. C. Hein et a/., Mol. Microbiol. 18, 77 (1995): j. Calan and R. Haguenauer-Tsapis, EMBOj. 16, 5847 (1997); C. Marchal, R. Haguenauer-Tsapis, D. Urban-Crimal, Mol. Cell. 5/01. 18, 314 (1998); B. Nefsky and D. Beach, EMBO 1. 15, 1301 (1996). 12. M. B. Yaffe et a/., Science 278, 1957 (1997); M. Schutkowski et a[., Biochemistry 37, 5566 (1998). 13. M. Shen, P. T. Stukenberg, M. W. Kirschner, K. P. Lu, Cenes Dev. 12, 706 (1998). 14. D. G. Crenshaw et al., EMBO j. 17, 1315 (1998). 15. Pinl and its mutants were produced, and their bind- ing to phosphoproteins was assayed, as described (72, 73). 16. Cdc25C was synthesized by in vitro transcription and translation in the presence of 35S-Met and phos- phorylated by Xenopus interphase or mitotic ex- tracts, and some mitotically phosphorylated samples were dephosphorylated with calf intestine phospha- tase (CIP) before binding, as described (73). 17. X. Z. Zhou, M. Schutkowski, G. Fisher, K. P. Lu, un- published data. 18. For peptide competition experiments, various pep- tides were incubated with CST-Pin1 or - W W domain in a binding buffer (73). After 1 hour of incubation, 32P-Labeled (27) or nonlabeled mitotic cell extracts were added. Proteins associated with glutathione beads were detected by autoradiography or immu- noblotting with MPM-2. To obtain semi-quantitative data, we scanned films of immunoblots at the region of 55 kD, the major Pinl-binding protein, and ana- lyzed data with ImageQuan (Scanjet II CX). Peptide dissociation constants were measured with a fluores- cence polarization assay as described (22). Peptides were labeled at the NH,-terminus and purified by thin-layer chromatography. Various concentrations of W W domain proteins were incubated with 0.1 p M of the Labeled peptides in a binding buffer containing 50 mM Hepes (pH 7.4), 100 mM NaCL, and 2% glycerol, and fluorescence polarization values were obtained with a PanVera 2000 system. 19. HeLa cells were labeled overnight with 32P-or- thophosphate or 35S-Met as described (27). Cells were lysed in Lysis buffer with or without phospha- tase inhibitors (40 mM glycerol phosphate, 50 mM NaF, 10 m M NaVO,, and 2 p M okadeic acid) (72, 73). For dephosphorylation experiments, three Ser phophatases (CIP, PPI, and PP2A) were added to Lysates for 30 min at 30°C in the absence or presence of the above phosphatase inhibitors as described (27). 20. j. janin and C. Chothia, j. Biol. Chem. 265, 16027 (1990): R. R. Copley and G. J. Barton, 1. Mol. 5/01, 242, 321 (1994); T. Clackson and J. A. Wells, Science 267, 383 (1995); !. Young, R. L. jernigan, D. C. Covell, Protein Sci. 3, 717 (1994). 21. P.-J. Lu, X. Z. Zhou, M. Shen, K. P. Lu, data not shown. 22. 1 . Radhakrishnan et a/., Cell 91, 741 (1997); D. Parker et a/., Mol. Cell 2, 353 (1998). 23. To determine the function of Pinl mutants in vivo, cDNAs of PIN7 or its mutants were subcloned into the yeast expression vector YEp451 and transformed into a temperature-sensitive ptfl strain YMP2, as described (28). Those strains expressing similar amounts of Pinl proteins were selected, as detected by immunoblotting with the 12CA5 antibody to the inserted NH2-terminal hemagglutinin A epitope tag (27). Three to four independent strains were tested for each transformation. 24. F. X. Schmid, Curr. Biol. 5, 993 (1995); T. Hunter, Cell 92, 141 (1998). 25. Z. Songyang et dl., Nature 373, 536 (1995); B, j. Mayer et al., Curr. Biol. 5, 296 (1995). 26. T. R. Coleman et al., Cell 72, 919 (1993); T. lzumi and J. L. Maller, Mol. Biol. Cell 4, 1337 (1993); ibid. 6, 215 (1995); N. Matsumoto-Taniura et al., ibid. 7, 1455 (1996); R. W. Kinget dl., Cell 81, 279 (1995); X. S. Ye et a/., EMBO]. 14, 986 (1995); A. Kumagai and W. C. Dunphy, Science 273, 1377 (1996); P. T. Stukenberg et a/., Curr. Biol. 7, 338 (1997). 27. K. P. Lu, 8. A. Osmani, A. R. Means.]. Biol. Chem. 268, 8769 (1993); K. P. Lu and T. Hunter, Cell 81, 413 (1995). 28. j. Hani, C. Stumpf, H. Domdey, FEBS Lett. 365, 198 (1995); S. D. Hanes, P. R. Shank, K. A. Bostian, Yeast 5, 55 (1989). 29. We thank two anonymous reviewers, T. Hunter, B. Neel, and L. Cantley for constructive comments, D. Parker and M. Montony for help with fluorescence polarization assay, and T. Stukenberg, M. W. Kirsch- ner, M. Sudol, j. Kunag, j. Hani, H. Domdey, M. Schutkowski, and C. Fischer for providing reagents. K.P.L. is a Leukemia Society of America Scholar. The work was supported by NIH grants ROlCM56230 and ROlGM58556 to K.P.L. 10 September 1998; accepted 27 January 1999 Facilitation of Signal Onset and Termination by Adenylyl Cyclase Klaus Scholich,' Jason 6. Mullenix,' Claus Wittpoth,' Helen M. Poppleton,' Sandra C. Pierre,' Margaret A. Lind~rfer,~ James C. Garrison,' Tarun 6. Patel'" The cr subunit (GSm) of the stimulatory heterotrimeric guanosine triphosphate binding protein (G protein) G, activates all isoforms of mammalian adenylyl cyclase. Adenylyl cyclase (Type V) and its subdomains, which interact with GSa, promoted inactivation of the G protein by increasing its guanosine triphos- phatase (GTPase) activity. Adenylyl cyclase and its subdomains also augmented the receptor-mediated activation of heterotrimeric G, and thereby facilitated the rapid onset of signaling. These findings demonstrate that adenylyl cyclase functions as a GTPase activating protein (GAP) for the monomeric GSm and enhances the GTPIGDP exchange factor (GEF) activity of receptors. Regulators of G protein signaling (RGS and onset of signals mediated by the GI and proteins) increase the GTPase activity of a G, families of G proteins (1-8). To date, subunits of heterotrimeric G proteins and however. no protein that acts as a specific play an important role in the termination GAP for Gbm has been identified. Because all 1328 26 FEBRUARY 1999 VOL 283 SCIENCE www.sciencemag.org